TW201421526A - Charged-particle radiation apparatus - Google Patents

Abstract

The present invention is a charged-particle radiation apparatus provided with a defect observation device for observing defects on a sample, the apparatus being provided with a control unit and a display unit, the control unit performing drift-correction processing according to a plurality of correction conditions on at least one image obtained by the defect-observation device, and the plurality of correction conditions and a plurality of corrected images on which drift-correction processing has been performed are associated and displayed on the display unit as a first screen.

Description

Charged particle beam device

The present invention relates to a charged particle beam device including a defect observation device of a semiconductor device.

In order to ensure high yields in semiconductor manufacturing, it is important to detect defects occurring in manufacturing engineering and implement countermeasures at an early stage. In recent years, with the miniaturization of semiconductors, the defects affecting the yield are also diversified, and the manufacturing engineering to be observed is also increasing. For example, more and more cases show that manufacturing engineering that may cause drift like the charging of the sample becomes the object of defect observation.

SEM (Scanning Electron Microscope) type defect observation device is a device for observing such various defects, generally for the defect position image detected by the defect inspection device of the previous position, with a defect than the previous position The inspection device is also of high quality for observation. Specifically, the defect coordinates outputted by the defect inspection device that moves the sample platform in front of the position are photographed at a low magnification to observe the extent that the defect of the object falls within the field of view to find the correct defect position, and then move the sample platform. Or move the shooting center to let the defect position come to the center of the field of view, and then The image for observation is obtained at a high magnification of defect observation. The reason why the defect position is ascertained by the low-magnification image as described above is that the defect coordinates outputted by the defect inspection device of the front-order position contain an error within the device specification range, so that the SEM-type defect observation device achieves high image quality. In the case of a defective image, there must be a process for correcting the error. The engineering automation technology for obtaining high-definition defect images is ADR (Automatic Defect Review/Redetection).

In the ADR, it is necessary to optimize the acquisition conditions of the low-magnification image and the acquisition conditions of the high-magnification image in order to balance the defect detection rate of the ADR with the defect coordinate detection accuracy or the sample characteristics of the defect inspection device of the previous position. ADR capacity including image acquisition time; but in general, ADR's defect detection rate and capacity are trade-off relationships, so even for experienced programmers, it is best to decide the best. The condition is still a difficult job, and it is expected to make the optimal condition setting work easier.

In addition, the automatic technology (ADC), which is an automation technology for identifying defects, based on the defect image obtained with high image quality, has also begun to be put into practical use. Especially in the production line, the engineering using ADC has been gradually expanded. In the ADC, the defect classification accuracy of the ADC is also a trade-off relationship with the ADC capacity including the image acquisition time. Therefore, it is difficult to determine the optimal condition, and it is expected to make the optimal condition setting easy. .

Patent Document 1 discloses a technique in which a plurality of scanning observation fields are obtained in a scanning electron microscope. The image is framed, and the amount of drift between the image frames is calculated, and then the amount of drift is corrected and the frame image is superimposed, whereby a sharp image can be obtained even in the case of image drift.

[Previous Technical Literature] [Patent Document]

[Patent Document 1] International Publication No. 2010/070815

However, the technique of Patent Document 1 is directed to image drift in the case of performing automatic length measurement. In Patent Document 1, although the length measurement value in the high-magnification image can be stably calculated, when the defect observation device is used, the following problem occurs.

According to Patent Document 1, in the scanning electron microscope for the purpose of manufacturing the automatic length measurement of the pattern, the user sets the manufacturing pattern of the length measuring object according to each sample or process recipe, in a sample or process parameter. Inside, the type of manufacturing pattern of the length measuring object is limited. That is, in the automatic length measurement, the coordinates determined in advance are measured, and the predetermined manufacturing pattern is measured, so that the optimum parameters do not change, for example, between samples.

On the other hand, in the SEM type defect observation apparatus, since the defect position image detected by the defect inspection apparatus of the front position is obtained, the coordinates and the manufacturing pattern to be obtained may vary depending on the position of the defect. move. Therefore, even within the same sample or process parameters, the coordinates and manufacturing drawings to be obtained can be varied. As a result, in the acquired image, since the degree of image drift due to charging changes due to each manufacturing pattern or the like, the setting of the optimum parameter becomes a problem. In the past, manufacturing processes such as drift have rarely become defects. However, in recent years, due to the miniaturization of semiconductors and the complication of manufacturing engineering, it is necessary to observe SEM-based defects for manufacturing processes where image drift occurs. The device is used to obtain high-definition defect images and analyze defects.

The present invention has been developed in view of such a situation, and provides a technique for easily determining the optimum parameter condition of an observed image even in the case where image drift occurs in a charged particle beam device including a defect observation device. .

In order to solve the above problems, for example, the configuration described in the scope of the patent application is adopted. The present application includes a plurality of means for solving the above problems, and an example of the present invention provides a charged particle beam device including a defect observation device for observing defects on a sample, the charged particle beam device having: a control unit; and a control unit that performs a drift correction process on a plurality of correction conditions on the one or more images acquired by the defect observation device, and performs the plurality of correction conditions and the drift correction process The plurality of corrected images are associated and displayed on the display unit as the first screen.

According to the present invention, in the charged particle beam device including the defect observation device, even when the image drift occurs, the optimum parameter condition of the observed image can be easily determined.

Other features related to the present invention will be apparent from the description and the drawings. In addition, the problems, configurations, and effects other than the above will be apparent from the description of the embodiments below.

101‧‧‧Electronic gun

102‧‧‧ lens

103‧‧‧Scan deflector

104‧‧‧object lens

105‧‧‧ samples

106‧‧‧ platform

107‧‧‧One electron beam

108‧‧‧ secondary particles

109‧‧‧Secondary particle detector

110‧‧‧Electronic Systems Control Department

111‧‧‧A/D conversion department

112‧‧‧ Platform Control Department

113‧‧‧All Control Department and Analysis Department

114‧‧‧Image Processing Department

115‧‧‧Operation Department

116‧‧‧ memory device

117‧‧‧Optical microscope

201‧‧‧Operation/Analysis Department

202‧‧‧Defect Data Memory Department

203‧‧‧Image Data Memory Department

204‧‧‧Analytical parameter memory

205‧‧‧Analysis Results Data Memory Department

Fig. 1 is a schematic model view showing the overall configuration of a SEM type defect observation apparatus of the present invention.

FIG. 2 is a detailed schematic view of the entire control unit and the analysis unit of FIG. 1. FIG.

[Fig. 3] A conceptual diagram of the image drift correction.

Fig. 4 is a flow chart showing the process of setting the number of frames accumulated in the first embodiment.

[Fig. 5] An example of a GUI for optimizing the number of accumulated sheets of the frame image is the first example of the screen displayed in step 403 of Fig. 4 .

[Fig. 6] An example of a GUI for optimizing the number of accumulated sheets of the frame image is a second example of the screen displayed in step 403 of Fig. 4 .

[Fig. 7] A flow chart of a condition setting process in the second embodiment, which is a process flow setting process for both ADR defect detection rate and throughput.

[Fig. 8] An example of a GUI for setting the condition of the ADR defect detection rate and the capacity, and is an example of a screen displayed in step 704 of Fig. 7.

[Fig. 9] A flow chart of a condition setting process in the third embodiment, which is a process flow setting process that takes into consideration both the ADC classification accuracy rate and the throughput.

[Fig. 10] An example of a GUI for setting the conditions of the ADC classification accuracy and productivity, and is an example of a screen displayed in step 904 of Fig. 9.

[Fig. 11] A flowchart of the condition setting processing of the fourth embodiment, which is a flowchart for setting a condition that takes into consideration both the ADR defect detection rate and the throughput, and the ADC classification accuracy and productivity.

Embodiments of the present invention will be described below with reference to the drawings. In addition, the drawings are intended to be illustrative of the invention, and are not intended to limit the invention.

The charged particle beam device is a device that accelerates an electric field (charged particles) having a charge such as an electron or a cation in an electric field and irradiates the sample. The charged particle beam device uses the interaction between the sample and the charged particles to observe, analyze, and process the sample. Examples of the charged particle beam device include an electron microscope, an electron beam drawing device, an ion processing device, and an ion microscope. In these charged particle beam apparatus, a scanning electron microscope (SEM) scans electrons to a sample, and detects an interaction between an electron and a sample as a signal, thereby performing fine structure observation or elemental analysis. Device. Hereinafter, as an example of a charged particle beam apparatus including a defect observation apparatus, an SEM type defect observation apparatus will be described.

Hereinafter, a configuration example of the SEM type defect observation device will be described. As an example of the configuration of the system, an example in which the process parameter setting is performed in the SEM type defect observation device will be described. However, the system configuration is not limited thereto, and some or all of the devices constituting the system may be configured as different devices. For example, the process parameter setting process of this embodiment may also be performed by a process parameter management device or a defect automatic classification device that is connected to the SEM type defect observation device.

The SEM type defect observation device refers to a defect coordinate detected by a defect inspection device such as an optical or SEM type inspection device, or an observation obtained by simulation or the like according to design layout data. Point coordinate information is used as input information to obtain a high quality SEM image of a defect or observation coordinate, which is suitable for observation or analysis.

Fig. 1 is a schematic view showing the overall configuration of a SEM type defect observation apparatus of the present embodiment. The SEM type defect observation apparatus of FIG. 1 includes an electro-optical system including optical elements such as an electron gun 101, a lens 102, a scanning deflector 103, a counter lens 104, a sample 105, and a secondary particle detector 109.

Further, the SEM type defect observation apparatus includes a stage 106 for moving a sample stage holding the sample 105 as an observation target in the XY plane, and an electro-optical system control unit 110 for controlling various optical elements included in the electro-optical system; The A/D conversion unit 111 quantizes the output signal of the secondary particle detector 109, and the platform control unit 112 controls the platform 106. The electronic optical system, the electro-optical system control unit 110, the A/D conversion unit 111, the platform 106, and the platform control unit 112 is a scanning electron microscope (SEM) which is a means for capturing an SEM image. Further, the SEM type defect observation device may be provided with an optical microscope 117 as a defect inspection device in front of the position.

Further, the SEM type defect observation device includes the entire control unit and the analysis unit 113, the image processing unit 114, the operation unit 115, and the memory device 116. The operation unit 115 includes a display (display unit), a keyboard, a mouse, and the like. In the memory device 116, an image obtained by SEM is stored.

In the SEM type defect observation apparatus, a primary electron beam 107 emitted from the electron gun 101 converges on the lens 102 and is deflected at the scanning deflector 103. Further, after the scanning electron deflector 103 is deflected by the scanning deflector 103, the primary electron beam 107 converges on the objective lens 104 and is irradiated to the sample 105.

The sample 105 irradiated with the primary electron beam 107 generates secondary particles 108 such as secondary electrons or reflected electrons depending on the shape or material of the sample 105. The generated secondary particles 108 are detected by the secondary particle detector 109, and then converted into digital signals by the A/D conversion unit 111. The output signal of the secondary particle detector 109, which is converted into a digital signal, is sometimes referred to as an image signal.

The output signal of the A/D conversion unit 111 is output to the image processing unit 114, and the image processing unit 114 forms an SEM image. The image processing unit 114 performs a drift correction process using the generated SEM image. Further, the image processing unit 114 can also perform various kinds of image analysis processing using the generated SEM image, such as ADR processing for performing image processing such as defect detection, ADC processing for automatically classifying defects according to categories, and the like.

The control of the electron optical system such as the lens 102, the scanning deflector 103, and the objective lens 104 is performed by the electro-optical system control unit 110. Further, the position control of the sample 105 is performed by the platform 106 controlled by the platform control unit 112. The overall control unit and the analysis unit 113 are control units that collectively control the entire SEM type defect observation device, and process input information from the operation unit 115 and the memory device 116 including a display, a keyboard, a mouse, and the like, and control the electronic optical system control. The unit 110, the platform control unit 112, the image processing unit 114, and the like output the processing result to the display unit or the memory device 116 included in the operation unit 115 as necessary.

The image processing unit 114, the overall control unit, and the analysis unit 113 are configured by an information processing device such as a computer. For example, the overall control unit and the analysis unit 113 are constituted by a CPU, a storage unit (for example, a memory and a hard disk), and an operation unit 115 including a display, a keyboard, a mouse, and the like. In this case, the overall control unit and the analysis unit 113 can be realized by software, and can be realized by a program in which the CPU executes a required arithmetic process. Similarly, the image processing unit 114 can also be implemented by software. Further, the image processing unit 114, the overall control unit, and the analysis unit 113 may be configured by individual information processing devices, or may be configured by one information processing device.

Further, the processing executed by the image processing unit 114, the entire control unit, and the analysis unit 113 may be implemented in a hardware manner. In the case of hardware execution, a plurality of actuators performing processing can be accumulated on a wiring substrate, or a semiconductor wafer or a package, thereby realizing.

FIG. 2 shows a detailed view of the overall control unit and the analysis unit 113 of FIG. 1. The operation/analysis unit 201 shown in FIG. 2 is the whole of FIG. The body control unit, the analysis unit 113, and the operation unit 115 integrate and express the objects.

The operation/analysis unit 201 includes a defect data storage unit 202, an image data storage unit 203, an analysis parameter storage unit 204, and an analysis result data storage unit 205. The defect data storage unit 202, the image data storage unit 203, the analysis parameter storage unit 204, and the analysis result data storage unit 205 may be configured by a hard disk constituting the information processing device of the entire control unit and the analysis unit 113. Further, when the operation/analysis unit 201 is assembled in the SEM type defect observation apparatus shown in FIG. 1, the defect data storage unit 202, the image data storage unit 203, the analysis parameter storage unit 204, and the analysis result data storage unit 205, It can also be integrated in the memory device 116 of FIG.

The defect data storage unit 202 stores defect information such as defect coordinates detected by the inspection device of the previous order. The image data storage unit 203 stores a defective image captured by the SEM type defect observation device. Here, the defective image may include a low-magnification image captured by the defect inspection device and a high-magnification image after the ADR process. The analysis parameter storage unit 204 stores a plurality of execution conditions (a plurality of parameters) executed when image acquisition or image analysis is performed. Examples of a plurality of execution conditions include parameters such as the cumulative number of frames, the voltage value of the acceleration voltage, and the current value of the probe current. In addition, a plurality of execution conditions may also store parameters such as ADR conditions and ADC conditions. The analysis result data storage unit 205 stores the processing result data of the operation/analysis unit 201. For example, the analysis result data storage unit 205 stores an image processed by a plurality of execution conditions, or processing time or capacity information when processed by each execution condition.

The operation/analysis unit 201 executes a predetermined program by the CPUs incorporated in the entire control unit and the analysis unit 113 in response to an operation instruction from the operation unit 115. In this way, the operation/analysis unit 201 can implement a plurality of functions. For example, the operation/analysis unit 201 acquires the defect information from the defect data storage unit 202, and acquires the defect image from the image data storage unit 203. Next, the operation/analysis unit 201 acquires a plurality of execution conditions from the analysis parameter storage unit 204, and performs processing on the defective image for each execution condition. The operation/analysis unit 201 stores information such as an image after execution of the processing in the analysis result data storage unit 205.

Further, the configuration in which the entire control unit and the analysis unit 113 shown in FIG. 1 are incorporated in the SEM type defect observation device is not limited, and may be configured as shown in FIG. 2 independently of the SEM type defect observation device shown in FIG. 1 . Operation/analysis unit 201. In this case, the SEM type defect observation device and the operation/analysis unit 201 are connected, for example, through a network.

Figure 3 is a conceptual diagram of the image drift correction. Here, an example in which the drift correction is performed on the three first frame images 301, the second frame image 302, and the third frame image 303 in which the drift occurs will be described. First, the amount of drift of the first frame image 301 is calculated based on the second frame image 302, and the overlapping positions are shifted by the amount corresponding to the calculated amount of drift (304). Similarly, the amount of drift of the third frame image 303 is calculated based on the second frame image 302, and the overlapping positions are shifted by the amount corresponding to the calculated amount of drift (304). The result of superimposing the calculated drift amount and superimposing the three frame images is the frame cumulative image 304.

In the example of FIG. 3, the second frame image 302 is Although the amount of drift is calculated by the reference, the first frame image 301 obtained first may be used as a reference, and the process of calculating the amount of drift may be repeated between consecutive frame images. Further, here, as the final drift correction image 305, the common portion of the total of three frame images 301, 302, and 303 is cut out in the frame cumulative image 304, but the following processing may be applied. That is, the area where the required image size is insufficient is filled with a specific pixel value, or the area where the pixel value is calculated from the peripheral pixel value by image processing is used to fill the insufficient area. These image drift correction processing is executed by the image processing unit 114. Further, the image drift correction processing may be executed by the entire control unit and the analysis unit 113.

<First Embodiment>

Hereinafter, the optimization processing of the execution conditions of the first embodiment of the SEM type defect observation apparatus will be described. Fig. 4 is a flow chart showing the optimization process of the cumulative number of frames in the frame. Hereinafter, a process of optimizing the number of accumulated sheets of the frame image will be described as an example of the execution conditions. Here, the main body of the following processing is the overall control unit and the analysis unit 113.

In step 401, first, the overall control unit and the analysis unit 113 acquire a plurality of parameters regarding the cumulative number of frames in the analysis parameter storage unit 204. Next, the entire control unit and the analysis unit 113 acquire the maximum number of frame images to be evaluated from the image data storage unit 203.

In step 402, the overall control unit and the analysis unit 113 use the image processing unit 114 to change the cumulative number of frames in the acquired frame image (that is, to follow the obtained plurality of parameters) to perform the drift repair. Processing. The overall control unit and the analysis unit 113 store the execution result such as the drift correction processing in the analysis result data storage unit 205.

Next, in step 403, the overall control unit and the analysis unit 113 display the image of the cumulative number of frames and the image of the result of the drift correction processing in each of the cumulative number of sheets in the form of the correspondence relationship, and display them in the operation unit 115. Display unit (such as a display). The display screen of the operation unit 115 will be described later in detail.

Next, in step 404, the user selects the best image from among the displayed drift correction images. The overall control unit and the analysis unit 113 receive information of an image selected by the user via the operation unit 115. In this way, the optimum drift correction condition can be easily set.

Finally, in step 405, the overall control unit and the analysis unit 113 reflect the set drift correction condition on the process parameters stored in the memory device 116. In this way, it can be applied to the next observation of defects. By following such a flow chart, the user can easily set the optimum drift correction condition.

FIG. 5 is an example of a GUI (first screen) for optimizing the number of accumulated sheets of the frame image, and is a first example of the screen displayed in step 403 of FIG.

The GUI of FIG. 5 includes a frame total number of sheets display unit 501, an image display unit 502 before the correction processing of the integrated image before the correction processing, and an image display after the correction processing of the integrated image after the correction processing. The unit 503 and the processing time display unit 504 that displays the execution time of the correction processing.

In the frame cumulative number display unit 501, the number of accumulated frames is displayed, and the comparison evaluates the minimum number of frames to be evaluated, the number of sheets of the smallest frame, and the number of sheets of the smallest frame. Times. Further, the selection of the cumulative number of frames is not limited to this method, and may be a combination of the minimum value, the central value, and the maximum value, or may not be a fixed value, and can be arbitrarily set by the user. Further, the number of comparisons is not limited to three types, and the total number of frames to be evaluated may be displayed one by one, or the method of selecting the optimum values in stages may be used in a plurality of times.

The pre-correction processing image display unit 502 displays an integrated image for each accumulated number of sheets before the correction processing. As shown in FIG. 5, in the case where image drift occurs, by accumulating the image, the misalignment of the edge portion of the pattern contained in the image of the evaluation object is highlighted (thickened). In this example, since the image drift occurs, the cumulative number of sheets increases, and the misalignment of the edge portion of the pattern contained in the image of the evaluation object increases. Further, in the sample to be observed as a defect, there is also a sample that does not need to perform the drift correction processing. Therefore, by displaying the frame integrated image in which the drift correction processing has not been performed, it is possible to determine whether or not the drift correction processing is necessary.

Further, after the correction processing, the image display unit 503 displays the integrated image after the correction processing for each accumulated number of sheets. By the correction processing, the misalignment of the edge portion of the pattern in each accumulated number of sheets becomes small. In this way, the image before the correction processing and the image after the correction processing corresponding to the total number of sheets of each frame are displayed in a form that can be associated with the cumulative number of frames.

Further, the processing time display unit 504 is a shape that can understand the correspondence relationship with the number of sheets of each frame and the integrated image of each frame. The formula shows the drift correction processing time. The user can easily select the optimum condition from the combination of the drift correction image (503) and the processing time required for the drift correction processing (504), which actually performs the drift correction processing. In the example of Fig. 5, an image with an accumulated number of sheets of 8 sheets (505) is selected. After selecting the best image and pressing button 506, the selected optimal condition (here, the cumulative number of sheets = 8) is saved to the process parameters as the next drift correction condition.

Hereinafter, another example of the screen displayed in the cumulative number-of-sheets optimization setting of the frame image will be described. FIG. 6 is an example of a GUI (second screen) for optimizing the number of accumulated sheets of the frame image, and is a second example of the screen displayed in step 403 of FIG.

In the SEM type defect observation apparatus, when the defect coordinates detected by the defect inspection apparatus of the front position are observed, since the manufacturing pattern of the defect is various, it is important to perform the condition setting system in response to various manufacturing patterns. . Ideally, if one parameter is valid for a wide variety of manufacturing drawings, then this parameter can be used; but such parameters, on the other hand, are mostly longer processing times. The user must consider the balance between the processing time and the processing time to set the optimal parameters, which is a difficult task.

Fig. 6 is a graph showing the results of optimizing the cumulative number of frames in the example of Fig. 5 for a plurality of evaluation samples, expressed as cumulative degrees. In the graph of FIG. 6, the horizontal axis represents the cumulative number of frames 601, the vertical axis (left) represents the cumulative degree 602, and the vertical axis (right) represents the drift correction processing time 603. In addition, in the graph of Fig. 6, the total number of sheets in each frame is depicted. The average drift corrects the processing time and displays an approximate line 605 of the plotted point. In the example of Fig. 6, the correction processing time is displayed by a straight line approximation. However, depending on the correction processing algorithm, it may not be displayed in a straight line. In this case, the curve may be displayed as an approximate curve, as long as it can be confirmed for a plurality of It is sufficient to evaluate the cumulative number of samples and the drift correction processing time in the total number of sheets of each frame.

By displaying the graph in this way, for a plurality of evaluation samples, the user can comprehensively confirm the result judged to be the best. For example, it is possible to determine the cumulative number of sheets of the frame that the user can satisfy the image quality for all the evaluation samples from the cumulative number of frames in which the cumulative degree is 100% (604). In the example of Fig. 6, in the case where the cumulative number of frames is 14, the cumulative degree becomes 100%. Therefore, as long as the cumulative number of frames is set to 14, the user can obtain the image quality judged to be the best for all the evaluation samples. In addition, at this time, it can be confirmed that the drift correction processing time is about 300 ms.

Further, when the drift correction processing time has an upper limit, for example, the drift correction processing time 606a, 606b as the upper limit can be displayed. The upper limit time can be set in advance or can be designed to be arbitrarily input by the user. For example, 606a is a case where the upper limit time of the drift correction processing time as the upper limit is set to 350 ms. In this case, the drift correction processing time of all the accumulated frame numbers is less than the upper limit time, so the user only needs to select the cumulative number of frames in which the cumulative degree becomes 100%.

For example, 606b is a case where the upper limit time of the drift correction processing time as the upper limit is set to 250 ms. It is known that this upper limit is met. The time is the number of frames with a cumulative number of 12 or less. Here, when the cumulative number of frames is 12, the cumulative degree is 95%, and it can be confirmed that the image quality judged by the user is about 95%. It can be seen that if the user sets the cumulative number of frames to 12 in the upper limit time, the image quality that is almost satisfied by the user can be obtained. In this way, it is possible to select both the cumulative degree 602 and the drift correction processing time 603 for each accumulated number of frames, and select the optimal parameter. When the user selects the optimal number of frames and presses the button 608, the selected optimal condition (here, the cumulative number of sheets = 14) is saved to the process parameters as the next drift correction condition.

In the above example, as an example of the execution condition, the process of optimizing the number of frames to be integrated is described. However, the execution condition (parameter) to be optimized is not limited to the number of frames to be accumulated. As described above, as the execution condition, parameters such as the voltage value of the acceleration voltage and the current value of the probe current can be optimized. In this case, the condition for a plurality of accelerating voltages displays an image on the display portion, and the user selects an optimum accelerating voltage condition. In addition, when a condition such as an acceleration voltage is set, it can be performed before the optimization process of the number of frames accumulated.

According to the present embodiment, the drift correction processing is performed on a plurality of drift correction conditions (the cumulative number of frames), and the plurality of drift correction conditions and the drift correction image subjected to the correction processing by the plurality of drift correction conditions are associated with each other. display. Therefore, even in the case where the image drift occurs, the optimum correction condition of the observed image can be easily determined. In addition, even in the case where the optimum drift correction condition varies depending on each evaluation sample due to the diversity of the manufacturing pattern to be observed, the user can still easily Set the best conditions.

<Second embodiment>

Hereinafter, the optimization processing of the execution conditions of the second embodiment of the SEM type defect observation apparatus will be described. The second embodiment is an optimum observation condition setting process for both the defect detection rate and the throughput of the automatic defect improvement (ADR) (Automatic Defect Review/Redetection). The ADR process corrects the error of the defect coordinates outputted by the defect inspection device of the previous order, detects the defect area, the defect coordinates, and the like, and obtains a high-definition defect image. Fig. 7 is a flow chart showing a condition setting process for both the defect detection rate and the capacity of the ADR. Hereinafter, a process of optimizing the number of accumulated sheets of the frame image will be described as an example of the execution conditions.

In the SEM type defect observation device, when the image of the defect coordinate detected by the defect inspection device of the front position is automatically captured by the ADR, the defect detection coordinate accuracy of the defect inspection device is considered, firstly, the defect is included in the field of view. The image is acquired at a low magnification, and the obtained low-magnification image is used for defect detection, and then the high-magnification image of high image quality is obtained so that the detected defect coordinates become the center of the field of view. Low-magnification images for defect detection are more important than ADR for defect detection systems. Therefore, whether ADR can correctly detect the defect position in the low-magnification image of ADR for defect detection is an important indicator for parameter setting.

In general, in the cumulative image of the frame, as the cumulative number of sheets increases, the noise component decreases, so from the viewpoint of the defect detection rate, The more the cumulative number of sheets is, the better it is. However, if the cumulative number of sheets is increased, the processing time will also increase. In particular, in the case where the drift correction processing is performed, it is necessary to have a process of calculating the amount of drift between the respective frame images, so that an increase in the processing time required for the accumulation of the frames causes a problem. Under such conditions, it is necessary to consider the balance between the ADR defect detection rate and the ADR capacity including the frame cumulative processing time to set the optimal conditions, which is a difficult setting item in the process parameter setting.

Hereinafter, the flowchart of Fig. 7 will be described. Here, the main body of the following processing is the overall control unit and the analysis unit 113.

In step 701, first, the overall control unit and the analysis unit 113 acquire a plurality of parameters relating to the cumulative number of frames from the analysis parameter storage unit 204. Next, the entire control unit and the analysis unit 113 acquire the maximum number of frame images to be evaluated from the image data storage unit 203.

Next, in step 702, the overall control unit and the analysis unit 113 use the image processing unit 114 to perform a change in the cumulative number of frames in the acquired frame image (that is, in accordance with the plurality of parameters obtained). Drift correction processing.

Next, in step 703, first, the overall control unit and the analysis unit 113 perform ADR processing on each of the frame integrated images before the drift correction processing. Further, the overall control unit and the analysis unit 113 perform ADR processing on each of the frame-accumulated images after the drift correction processing is performed to change the number of frames. Next, the overall control unit and the analysis unit 113 store the execution result of the ADR process in the analysis result data storage unit 205.

Next, in step 704, the overall control unit and the analysis unit 113 convert the drift correction image of the number of frames accumulated, the defect position detected by the ADR to each of the drift correction images, and the ADR capacity of each of the drift correction images. The display unit (such as a display) displayed on the operation unit 115 is displayed in a form in which the correspondence can be understood. The display screen of the operation unit 115 will be described later in detail.

Next, in step 705, the user selects the best image from among the displayed drift correction image and the ADR execution result. The overall control unit and the analysis unit 113 receive information of an image selected by the user via the operation unit 115. In this way, the optimum drift correction condition after considering the ADR can be easily set.

Finally, in step 706, the overall control unit and the analysis unit 113 reflect the optimal drift correction condition after the consideration of ADR on the process parameters stored in the memory device 116. In this way, it can be applied to the next observation of defects. By following such a flowchart, the user can easily set the optimal drift correction condition after considering the ADR.

FIG. 8 is an example of a GUI for setting the condition of the ADR defect detection rate and the capacity, and is an example of a screen displayed in step 704 of FIG.

The GUI of FIG. 8 includes a display selection unit 801 having an ADR result, a frame total number of sheets display unit 802, and an image display unit 803 before the correction processing of the integrated image before the display correction processing, and an accumulation after the display correction processing. The image display unit 806 after the image correction processing, the processing time display unit 807 that displays the execution time of the correction processing, and the capacity display unit 808 that displays the ADR capacity.

The display selection unit 801 of the ADR result selects whether or not to superimpose the ADR results together. When the check is made, the ADR result (the defective area 804 and the defective coordinates 805) are displayed in a superimposed manner on the image. In the example of FIG. 8, the defect area 804 detected by the ADR is displayed as a result of polygon grouping, but all of the detected defect areas may be overlaid and displayed without clustering. In addition, in the example of FIG. 8, the defect coordinate 805 detected by the ADR is the center of gravity of the defect area 804, but for example, a defect characteristic such as a luminance value (such as luminance) may be defined, and the most likely defect is determined. The pixel is used as the defect coordinate as long as the definition corresponding to the ADR defect detection algorithm is adopted.

The total number of sheets of the frame is displayed on the frame cumulative number display unit 802, and the comparison is performed to estimate the minimum number of frames to be evaluated, the number of sheets of the smallest frame, and the number of sheets of the smallest frame. Times. Further, the selection of the cumulative number of frames is not limited to this method, and may be a combination of the minimum value, the central value, and the maximum value, or may not be a fixed value, and can be arbitrarily set by the user. Further, the number of comparisons is not limited to three types, and the total number of frames to be evaluated may be displayed one by one, or the method of selecting the optimum values in stages may be used in a plurality of times.

Before the correction processing, the image display unit 803 displays an integrated image for each accumulated number of sheets before the correction processing. In the case where image drift occurs, by accumulating the image, the misalignment of the edge portion of the pattern or the defect contained in the image of the evaluation object is conspicuous (thickening) display. In the example of Fig. 8, due to the drift and noise components, if the cumulative number of sheets is less, the number is missing. The trapped area 804 is detected as wider. As a result, when the cumulative number of frames is 4 or 8, the defective coordinates 805 are deviated and detected relative to the defect position 811. Further, in the sample to be observed as a defect, there is also a sample that does not need to perform the drift correction processing. Therefore, by displaying the frame integrated image in which the drift correction processing has not been performed, it is possible to determine whether or not the drift correction processing is necessary.

Further, after the correction processing, the image display unit 806 displays the integrated image after the correction processing for each accumulated number of sheets. As shown in FIG. 8, the pre-correction processed image and the corrected processed image corresponding to the cumulative number of frames in each frame are displayed in a form that can be associated with the cumulative number of frames. Further, by the correction processing, the misalignment of the edge portion of the pattern or the defect in each of the cumulative number of sheets becomes small, and as a result, the defective area 804 becomes smaller than before the correction processing. As a result, the deviation of the defective coordinates 805 becomes smaller with respect to the defect position 811.

Further, the drift correction processing time is displayed on the processing time display unit 807 in a form in which the correspondence relationship between the accumulated number of sheets of each of the frames and the integrated image of each of the frames in which the ADR result is displayed is known. In addition, the ADR capacity including the drift correction processing time is displayed on the capacity display unit in a form that can understand the correspondence relationship between the accumulated number of frames and the integrated image of each frame in which the ADR result is displayed. 808. The processing time discussed in the case of using ADR is often the ADR capacity including the drift correction processing time. Therefore, it is desirable to mark not only the drift correction processing time (807) but also the ADR capacity (808).

Like this, as long as the GUI is used, the user can actually The optimal condition (809) is easily selected among the combination of the image after the drift correction processing, the ADR result for each drift correction image, and the ADR capacity for each drift correction image. After the best image is selected and the button 810 is pressed, the selected optimal condition (here, the cumulative number of sheets = 8) is saved to the process parameter as the drift correction condition after the next ADR consideration.

Further, in step 704 of FIG. 7, the same content as that of the screen illustrated in FIG. 6 may be displayed. The drift correction condition optimization operation after considering ADR for a plurality of samples can correspond to the content of the example of FIG. 6. In Fig. 6, the cumulative degree of the image judged to be the best in Fig. 5 is displayed. On the other hand, in the case where the drift correction condition after the ADR is optimized, it is only necessary to display the cumulative degree of the image determined to be the best after the user considers the ADR in FIG. In addition, in this embodiment, it is not limited to the cumulative degree, and other information may be displayed. For example, if the user judges that the image is the best, it means that the defect coordinates can be detected correctly, so the defect coordinates of the image of the image that is judged to be the best image by the user and the defect coordinates of the image of the other frame are made. By comparison, the detection rate of the defect can be calculated for each frame cumulative number. In this case, the detection rate for the cumulative number of each frame is also displayed in the chart. In addition, in FIG. 6, the drift correction processing time is displayed as a second axis, but in the case where the drift correction condition after the ADR is optimized, the ADR capacity including the drift correction processing time can be displayed as a graph. .

According to the present embodiment, the optimum drift correction condition varies depending on each sample even in view of the diversity of the manufacturing pattern to be observed. In the case of the situation, the drift correction condition after considering the ADR can still be easily selected. Further, in the setting process, the ADR capacity including the drift correction processing time is displayed, so that it is possible to easily set the conditions for both the ADR defect detection rate and the throughput.

<Third embodiment>

Hereinafter, the optimization processing of the execution conditions of the third embodiment in the SEM type defect observation apparatus will be described. The third embodiment is an optimum observation condition setting process for classifying the correct rate and productivity of the automatic defect classification (ADC: Automatic Defect Classification). The ADC process classifies defect types (identify defect types) based on defective images obtained with high image quality. Fig. 9 is a flow chart showing the process of setting the conditions for the classification accuracy and productivity of the ADC. Hereinafter, a process of optimizing the number of accumulated sheets of the frame image will be described as an example of the execution conditions.

In order to ensure the correct rate in the ADC, the defect must be analyzed with high image quality, and the object of the ADC, that is, the acquisition condition of the high-magnification image is very important. Depending on the algorithm of the ADC, sometimes not only high-magnification images but also low-magnification images, but the description here assumes that the image that has a large influence on the correctness of the ADC is a high-magnification image.

In general, in the integrated image of the frame after the drift correction, as the cumulative number of sheets increases, the noise component decreases. Therefore, from the viewpoint of the classification accuracy of the ADC, the more the cumulative number of sheets, the better, but If the cumulative number of sheets is increased, the ADC processing time including the drift correction processing time will also increase. In addition, it is suitable for the user to visually classify the image quality, and The image quality in the ADC that can obtain a sufficient positive solution rate is not necessarily the same. Therefore, it is necessary to consider the balance between the ADC classification correct rate and the cumulative number of frames and the ADC processing time including the drift correction processing time to set the optimal conditions. . Therefore, the conditions for optimizing the classification accuracy and productivity of the ADC are optimized, and it is a difficult operation in the process parameter setting.

Hereinafter, the flowchart of Fig. 9 will be described. Here, the main body of the following processing is the overall control unit and the analysis unit 113.

In step 901, first, the overall control unit and the analysis unit 113 acquire a plurality of parameters relating to the cumulative number of frames from the analysis parameter storage unit 204. Next, the entire control unit and the analysis unit 113 acquire the maximum number of frame images to be evaluated from the image data storage unit 203.

Next, in step 902, the overall control unit and the analysis unit 113 use the image processing unit 114 to perform a change in the number of accumulated frames (that is, in accordance with the plurality of parameters obtained) with respect to the acquired frame image. Drift correction processing.

Next, in step 903, first, the overall control unit and the analysis unit 113 perform ADC processing on each of the frame-accumulated images before the drift correction processing. Further, the overall control unit and the analysis unit 113 perform ADC processing on the image integration images after the drift correction processing is performed to change the cumulative number of frames. Next, the overall control unit and the analysis unit 113 store the execution result of the ADC processing in the analysis result data storage unit 205.

Next, in step 904, the overall control unit and the analysis unit 113 correct the drift correction image of each frame and the ADC for each drift correction. The classification result of the positive image and the ADC capacity of each of the drift correction images are displayed on the display unit (such as a display) of the operation unit 115 in a form in which the correspondence can be understood. The display screen of the operation unit 115 will be described later in detail.

Next, in step 905, the user selects the best image from among the classification results of the drift correction image and the ADC displayed. The overall control unit and the analysis unit 113 receive information of an image selected by the user via the operation unit 115. In this way, the optimum drift correction condition after considering the ADC can be easily set.

Finally, in step 906, the overall control unit and the analysis unit 113 reflect the optimal drift correction condition after the ADC is considered, on the process parameters stored in the memory device 116. In this way, it can be applied to the next observation of defects. By following such a flow chart, the user can easily set the optimum drift correction condition after considering the ADC.

FIG. 10 is an example of a GUI for setting the condition of the ADC classification accuracy and productivity, and is an example of a screen displayed in step 904 of FIG.

The GUI of FIG. 10 includes a display selection unit 1001 with an ADC result, a frame count display unit 1002, and a pre-correction image display unit 1003 for displaying an integrated image before the correction processing, and a display before the correction processing. The first ADC result display unit 1004 of the ADC result of the integrated image, and the image display unit 1006 after the correction processing of the integrated image after the correction processing is displayed, and the second ADC result of the ADC result of the integrated image after the correction processing is displayed. The display unit 1007 and the processing time display unit 1008 that displays the execution time of the correction processing and the display ADC capacity Capacity display unit 1009.

The display result selection unit 1001 of the ADC result selects whether or not to superimpose the ADC results together. When the check is made, the ADC result (the defective area 1005, the first ADC result display unit 1004, and the second ADC result display unit 1007) is displayed. In the example of FIG. 10, the defective area 1005 detected by the ADC is displayed as a result of polygon grouping, but all the defective areas detected may be overlaid without clustering.

In the frame cumulative number display unit 1002, the total number of frames is displayed, and the comparison is performed to estimate the minimum number of frames to be evaluated, the number of sheets of the smallest frame, and the number of sheets of the smallest frame. Times. Further, the selection of the cumulative number of frames is not limited to this method, and may be a combination of the minimum value, the central value, and the maximum value, or may not be a fixed value, and can be arbitrarily set by the user. Further, the number of comparisons is not limited to three types, and the total number of frames to be evaluated may be displayed one by one, or the method of selecting the optimum values in stages may be used in a plurality of times.

The pre-correction processing image display unit 1003 displays an integrated image for each accumulated number of sheets before the correction processing. In the case where the image drift occurs, by accumulating the image, the misalignment of the edge portion of the pattern or the defect contained in the image of the evaluation object is highlighted (bright). In the example of Fig. 10, due to the drift and noise components, the defect area 1005 is detected to be wider as the cumulative number of sheets is smaller. Further, in the sample to be observed as a defect, there is also a sample that does not need to perform the drift correction processing, so that it is possible to judge by displaying the integrated image of the frame in which the drift correction processing has not been performed. Whether to do drift correction processing.

The first ADC result display unit 1004 displays an ADC result for the cumulative image before the correction process for each accumulated number of sheets. In the example of Fig. 10, when the cumulative number of frames is 4, the classification result of the ADC cannot be determined and is displayed as "Unknown". When the total number of frames in the frame is 8, it is classified as "Short" (short circuit), and in this example, it does not get the correct classification result. In addition, when the cumulative number of frames is 16, it is classified as "Dust" (foreign matter), and the correct classification result is obtained.

Further, after the correction processing, the image display unit 1006 displays the integrated image after the correction processing for each accumulated number of sheets. As shown in FIG. 10, the pre-correction processed image and the corrected processed image corresponding to the cumulative number of frames in each frame are displayed in a form that can be associated with the cumulative number of frames. Further, by the correction processing, the misalignment of the edge portion of the pattern or the defect in each of the cumulative number of sheets becomes small, and as a result, the defective area 1005 becomes smaller than before the correction processing.

The second ADC result display unit 1007 displays an ADC result for the cumulative image after the correction processing for each accumulated number of sheets. In the example of Fig. 10, when the cumulative number of frames is 4, the classification result of the ADC cannot be determined even after the correction processing, and the display is "Unknown". When the cumulative number of frames is 8, it is classified as "Dust" (foreign matter), and the correct classification result is obtained before the correction processing. In addition, when the cumulative number of frames is 16, it is classified as "Dust".

Further, the drift correction processing time is displayed on the processing time display unit 1008 in a form in which the correspondence relationship between the cumulative number of sheets of each of the frames and the integrated image of each of the frames in which the ADC result is displayed is known. In addition, the ADC production capacity including the drift correction processing time is displayed on the capacity display unit in a form that can understand the correspondence relationship between the accumulated number of frames and the integrated image of each frame in which the ADC result is displayed. 1009. When the processing time in the case of using the ADC is often the ADC capacity including the drift correction processing time, it is desirable to not only mark the drift correction processing time (1008), but also the ADC capacity (1009). In addition, ADC processing is often side-by-side with ADR processing for pipeline processing. In particular, depending on the number of samples to be processed, the productivity of the ADC and the ADR may be the same. Therefore, the capacity of the ADC and the ADR are not specifically displayed here. However, if the ADC capacity is to be correctly identified, the capacity difference between the ADC and the ADR can also be displayed.

In this way, by using the GUI, the user can easily select from the combination of the image after the actual drift correction processing, the ADC result for each drift correction image, and the ADC capacity of each drift correction image. The best condition (1010). After selecting the best image and pressing button 1011, the selected optimal condition (here, the cumulative number of sheets = 8) is saved to the process parameter as the drift correction condition after the ADC is considered next time.

Further, in step 904 of FIG. 9, the same content as that of the screen illustrated in FIG. 6 may be displayed. The drift correction condition optimization operation after considering the ADC for a plurality of samples can be as the example of FIG. should. In Fig. 6, the cumulative degree of the image judged to be the best in Fig. 5 is displayed. On the other hand, in the case where the drift correction condition after the ADC is optimized is considered, it is only necessary to display the cumulative degree of the image judged to be the best after the user considers the ADC in FIG. In addition, in this embodiment, it is not limited to the cumulative degree, and other information may be displayed. For example, the image that the user judges to be the best image refers to an image that can correctly perform the defect classification image, so the user judges the ADC image of the best image and the image of the accumulated number of other frames. The results are compared so that the correct rate of defect classification can be calculated for each frame count. In this case, the defect classification correct rate for each frame count is also displayed in the graph. In addition, in FIG. 6, the drift correction processing time is displayed as a second axis, but in the case where the drift correction condition after the ADC is optimized, the ADC capacity including the drift correction processing time can also be displayed as a graph. .

According to the present embodiment, even in the case where the optimum drift correction condition varies depending on each sample due to the diversity of the manufacturing pattern to be observed, the drift correction condition after the ADC can be easily selected. In addition, during the setting process, the ADC capacity including the drift correction processing time is displayed, so that it is possible to easily set the conditions for the accuracy and productivity of the ADC defect classification.

<Fourth embodiment>

Hereinafter, the optimization processing of the execution conditions of the fourth embodiment of the SEM type defect observation apparatus will be described. The fourth embodiment is about achieving the ADR defect detection rate and productivity, and taking into account the accuracy of ADC classification and productivity. Check the condition setting process. Fig. 11 is a flow chart showing the process of setting conditions for achieving both ADR defect detection rate and throughput, and taking into account the accuracy and capacity of the ADC classification. Hereinafter, a process of optimizing the number of accumulated sheets of the frame image will be described as an example of the execution conditions.

Fig. 11 is a view in which the flow of the example of Fig. 7 and the flow of the example of Fig. 8 are integrated. Here, the main body of the following processing is the overall control unit and the analysis unit 113.

In step 1101, first, the entire control unit and the analysis unit 113 acquire a plurality of parameters relating to the cumulative number of frames from the analysis parameter storage unit 204. Next, the entire control unit and the analysis unit 113 acquire the maximum number of frame images to be evaluated from the image data storage unit 203.

Next, in step 1102, the overall control unit and the analysis unit 113 use the image processing unit 114 to perform a change in the number of frames accumulated (that is, in accordance with the plurality of parameters obtained) with respect to the acquired frame image. Drift correction processing.

Next, in step 1103, first, the overall control unit and the analysis unit 113 perform ADR processing on each of the frame integrated images before the drift correction processing. The overall control unit and the analysis unit 113 performs ADR processing on each of the frame-accumulated images after the drift correction processing is performed to change the cumulative number of frames. Next, the overall control unit and the analysis unit 113 store the execution result of the ADC processing in the analysis result data storage unit 205.

Next, in step 1104, the overall control unit and the analysis unit 113 convert the drift correction image of the number of frames accumulated, the defect position detected by the ADR to each of the drift correction images, and the respective drift correction images. The ADR capacity is displayed on the display portion (such as a display) of the operation unit 115 in a form capable of understanding the correspondence. The screen of Fig. 8 is displayed here.

Next, in step 1105, the user selects the best image from among the displayed drift correction image and the ADR execution result. The overall control unit and the analysis unit 113 receive information of an image selected by the user via the operation unit 115.

Next, in step 1106, first, the overall control unit and the analysis unit 113 perform ADC processing on each of the frame-accumulated images before the drift correction processing. Further, the overall control unit and the analysis unit 113 perform ADC processing on the image integration images after the drift correction processing is performed to change the cumulative number of frames. Next, the overall control unit and the analysis unit 113 store the execution result of the ADC processing in the analysis result data storage unit 205.

Next, in step 1107, the overall control unit and the analysis unit 113 convert the drift correction image of the number of frames in each frame, the classification result of the ADC to each of the drift correction images, and the ADC capacity of each of the drift correction images. The form of the correspondence relationship is known to be displayed on the display portion (such as a display) of the operation unit 115. The screen of Fig. 10 is displayed here.

Next, in step 1108, the user selects the best image from among the classification results of the drift correction image and the ADC displayed. The overall control unit and the analysis unit 113 receive information of an image selected by the user via the operation unit 115.

Finally, in step 1109, the overall control unit and the analysis unit 113 will consider the optimal drift correction condition after ADR, and the optimal drift correction condition after considering the ADC, reflected in the process parameters stored in the memory device 116.

By following such a flow chart, the user can easily set the optimum conditions for taking into consideration the ADR defect detection rate and productivity, and the optimum conditions for both the ADC classification accuracy and the throughput.

According to this embodiment, the drift correction conditions of both the ADR and the ADC can be continuously set. As described above, since the capacity of the ADC and the ADR may not be equal, for example, in step 1107, the capacity of the ADC and the ADR may be displayed differently. In this way, it is possible to determine the optimum drift correction conditions while comparing the capacity of both the ADC and the ADR. Further, by continuously setting the drift correction conditions of both the ADR and the ADC, the drift correction image processed in step 1102 can be directly used in step 1106, and the processing time can be shortened.

Further, the present invention is not limited to the above embodiments, and various modifications are also included. For example, the above-described embodiments are described in detail for the convenience of the description of the present invention, and are not necessarily limited to all of the configurations described. Further, a part of the configuration of a certain embodiment may be replaced with a configuration of another embodiment, and a configuration of another embodiment may be added to the configuration of a certain embodiment. Further, other configurations may be added, deleted, or replaced for a part of the configuration of each embodiment.

Further, as described above, the overall control unit, the analysis unit 113, and the image processing unit 114 may be realized by a software program code that realizes the functions of the embodiment. In this case, it can also be designed as a memory medium for recording code. Provided to the information processing device, the information processing device (or CPU) reads the code stored in the memory medium. In this case, the code itself read from the memory medium implements the functions of the foregoing embodiments, and the code itself, and the memory medium in which it is stored, constitute the present invention. As a memory medium for supplying code like this, for example, a floppy disk, a CD-ROM, a DVD-ROM, a hard disk, a compact disk, a magneto-optic disc, a CD-R, a magnetic tape, and a non-volatile one can be used. Memory card, ROM, etc. In addition, the existing device can be upgraded by recording the recording medium with the program.

Further, it is also possible to design an OS (operation system) or the like operating on the information processing device to perform part or all of the actual processing in accordance with the instruction of the code, and to implement the functions of the foregoing embodiments by the processing. In addition, the software code for implementing the functions of the embodiment may be designed to be distributed over the network and stored in a memory device of the information processing device or a memory medium such as a CD-RW or a CD-R. The CPU of the device reads the code stored in the memory device or the memory medium for execution.

The present invention has been described with respect to specific examples, but they are not intended to be limiting, but are intended to be illustrative. It will be apparent to those of ordinary skill in the art that, in order to practice the present invention, there are many combinations of hardware, software, and firmware. For example, the code for implementing the functions described in this embodiment can be a wide-area program or a script language such as an assembler, C/C++, perl, shell, PHP, or Java (registered trademark). Come and implement.

In addition, the control lines or information lines in the drawing reveal that all the control lines or information lines on the product are not necessarily disclosed. It can also be designed such that all the components are connected to each other.

501‧‧‧Drawing number of pictures

502‧‧‧Revised image display before processing

503‧‧‧Revision processed image display unit

504‧‧‧Processing time display

Claims (15)

A charged particle beam device comprising a defect observation device for observing a defect in a sample, the charged particle beam device comprising: a control unit; and a display unit; wherein the control unit is obtained by the defect observation device One or more images are subjected to drift correction processing under a plurality of correction conditions, and the plurality of correction conditions and a plurality of correction images subjected to the above-described drift correction processing are associated with each other and displayed on the display unit as The first screen. The charged particle beam device of claim 1, wherein the control unit performs an automatic defect observation process on the plurality of modified images, and superimposes the defect position detected by the defect automatic observation process to the foregoing A plurality of corrected images are displayed on the first screen. The charged particle beam device according to claim 2, wherein the control unit associates the capacity information of the defect automatic observation processing on the plurality of corrected images with the plurality of correction conditions, and displays the foregoing The first screen. The charged particle beam device of claim 2, wherein the control unit selects a distribution of the plurality of correction conditions selected by a user and a detection rate of the defect automatic observation processing for each of the plurality of correction conditions At least one is displayed on the display unit as the second Picture. The charged particle beam device of claim 4, wherein the control unit automatically performs an observation processing time of the drift correction processing on the one or more images and the defect automatic observation processing on the plurality of corrected images. At least one of the capacity information is displayed on the second screen. The charged particle beam device of claim 1, wherein the control unit performs defect automatic classification processing on the plurality of modified images, and classifies the classification result by the automatic defect classification processing, and the foregoing plural number The corrected images are displayed and displayed on the first screen. The charged particle beam device of claim 6, wherein the control unit associates the capacity information for the defect automatic classification processing on the plurality of corrected images with the plurality of correction conditions, and displays the foregoing The first screen. The charged particle beam device of claim 6, wherein the control unit automatically classifies the distribution of the plurality of correction conditions selected by a user and the defect of each of the plurality of correction conditions At least one of them is displayed on the display unit as a second screen. The charged particle beam device according to claim 8, wherein the control unit automatically classifies the execution time of the drift correction processing for the one or more images and the defect for the plurality of corrected images. At least one of the capacity information is shown in the second Picture. The charged particle beam device according to claim 1, wherein the control unit associates the execution time of the drift correction processing for the one or more images with the plurality of correction conditions, and displays the 1 screen. The charged particle beam device according to claim 1, wherein the control unit associates the one or more images before the drift correction processing with the plurality of correction conditions and displays the image on the first screen. The charged particle beam device according to claim 1, wherein the control unit displays the distribution of the plurality of correction conditions selected by the user on the display unit as the second screen. The charged particle beam device according to claim 12, wherein the control unit associates the execution time of the drift correction processing for the one or more images with the distribution of the plurality of correction conditions, and displays In the second screen above. The charged particle beam device according to claim 1, wherein the control unit performs defect automatic observation processing on the plurality of corrected images, and superimposes the defect position detected by the defect automatic observation processing to Displaying the plurality of modified images on the first screen, performing defect automatic classification processing on the plurality of modified images, and performing classification result obtained by the defect automatic classification processing, and the foregoing complex A plurality of corrected images are associated with each other and displayed on the display unit as a second screen. The charged particle beam device of claim 14, wherein the control unit automatically classifies the capacity information of the defect automatic observation processing for the plurality of modified images and the defect for the plurality of modified images. The capacity information is displayed on the second screen.